What is the difference between covariate and confounding variables? What do covariate and confounding variables have in common and how do they differ? And what are their specific effects in causal inference? (in statistics and causal inference)
 A: This is a complicated question because different fields conceive these types of variables differently, where others make no distinction whatsoever (which is the case for many social sciences fields and subfields).
In statistics, a confound is a variable that is so closely related or associated with another variable that you can’t tell their effects apart. In epidemiology, confounding variables to signify a covariate that is related to both predictors & treatment/exposure. 
There are also who focus on the effect of a confounder: "A Confounder is a variable whose presence affects the variables being studied so that the results do not reflect the actual relationship". Pourhoseingholi MA, Baghestani AR, Vahedi M. How to control confounding effects by statistical analysis. Gastroenterol Hepatol Bed Bench. 2012;5(2):79-83.
In practice, however, I have seen quite often the interchangeable use of covariates, confounding, predictor, & controls variables. I also seen the difference in nomenclature representing different theoretical considerations. For example, scholars A may classify variable X as a confounder while scholar B as covariate.
Either way, mathematically, and barring causal inferences, when estimating a linear regression or time-series, there is no difference. That is to say, it’s the same model & you run it the same way in statistical softwares. If you are interested in causal inferences, however, nomenclature start to matter and reflect different models of the phenomenon at hand.
